Fast Morphological Component Analysis Tangka Image Inpainting Based on Nonlocal Mean Filter

Thangka image contains rich content and complex color.In the context of image inpainting based on morphological component analysis,the imposition of a total variation penalty is useful,particularly well in recovering piecewise smooth objects,but global redundancy similar information is difficult to utilize effectively and easy to produce staircase.This paper proposed a new image inpainting method based on morphological component analysis.It utilized the information of pixels around fixed window to express the current pixel in order to preserve the fine structure and details and eliminated staircase simultaneously,which made the subsequent iteration more effective.In addition,the improved algorithm based on fast algorithm reduced the calculation weights of uncorrelated pixels within an area,thus decreased complexity of the algorithm.Experimental results for thangka image which contains scratch and block loss show that the proposed method achieves better inpainting effect.